Physical Sciences REU 2015 Awardees

Annette Hilton (The College of Wooster)

Mentor: Julianne Gross

Classifying The Unknown: The Lunar Edition

The composition of the lunar crust (and pristine lunar rocks) provides important clues about the processes that formed it and hence contains information on the origin and evolution of the Moon. The early history and broad-scale petrogenesis of the Moon’s evolution were glimpsed from the first Apollo sample return. However, these missions only sampled a limited part of the lunar nearside highlands, areas that are now known to be unusual in several respects (e.g., wide distribution of K-REE-P material). Thus, a fundamental property of the moon, the lithologies of its surface, is only partially known. Lunar meteorites come from random sites on the Moon--including areas not visited by Apollo or Lunamissions-- and therefore are crucial for understanding the evolution and development of the whole Moon. In this study we investigated an unclassified lunar meteorite, to 1) confirm its origin from the Moon, 2) classify the rock, 3) place constraints on its evolution, crystallization history (including pairing with other stones), and origin on the lunar surface area, and 4) improve our understanding of unsampled areas of the Moon and enlarge our knowledge of lunar highland rock types. We used microanalytical techniques such as the Electron Microprobe and Laser-Ablation Inductive-Coupled-Plasma Mass-Spectrometry to obtain chemical analyses of single mineral phases as well as whole rock areas. X-ray elemental mapping was used to calculate mineral abundances and ultimately the bulk rock chemistry. The Fe/Mn ratio of olivine and pyroxene in this rock falls within the ratio for known lunar samples, confirming its origin. The rock represents an anorthositic troctolite that contains mineral and rock fragments, as well as melt veins, set in a granulitic matrix. This variety of fragments as well as the texture of the sample suggests a complex formation history; it most likely originated within an impact crater near the lunar surface. The bulk rock has a high Mg# [= molar Mg/(Mg+Fe)] of  84 which is one of the highest Mg# ever reported for a lunar sample, thus making it a unique rock type. This high magnesian character of the sample suggest the presence of Mg-rich plutonic rocks, such as the Apollo Mg-Suite rocks, that was incorporated into the sample e.g. during an impact event. However, low the rare-earth-element abundances of the bulk rock and mineral grains indicate that the sample originated far away from the KREEP terrain and thus the Mg-suite rocks. A high magnesian mantle component, such as dunites, could serve as a possible constituent instead. However, this hypothesis remains to be tested.

Paco Holguin (MIT)

Mentor:  Ariyeh Maller

Comparing Simulations of Galaxy Halos

There is still no complete theory of galaxy formation and evolution.  One way of making progress is the use of hydrodynamical simulations, but there still remain many uncertainties in their use, particularly on scales not resolved. The goal of this project is to compare the results from five different hydrodynamical simulations of section of a universe at z=3.  We characterize these simulations by analyzing the relationship between total halo, star, and gas mass in dark matter halos.  We also compare these relationships with observational data.

Alexandra Mannings (U. Alabama)

Mentor: Mordecai MacLow/Jana Grcevich

Investigating The Cusp/Core Problem Through Simulations: Setting Up Dark Matter Initial Conditions

The purpose of the research is to investigate the effects of baryonic processes on the dark matter distribution in dwarf galaxies.  It is observed that the dark matter density profiles in dwarf galaxies is more of a shallow, cored profile.  This differs form the simulated profiles that have a steep, cusped profile.  Specifically, we are focusing on the Navarro, Frenk, and White (NFW) profile that produces accurate results on large scales but differs with observations at the level of dwarf galaxies.  I am writing a program that determines the distribution of dark matter particles so that they can be inserted into a simulation that now runs only with gas.  The code produces an energy distribution that will determine the energies at which particles can exist, and the radii and velocities corresponding to those energy values.  The ultimate goal is to insert the dark matter and see how the gas processes affect the distribution over the course of the simulation.

Ryan Flesch (College of William and Mary)

Mentor: Jim Webster/Patricia Nadeau

Save Your Apatite!  A Quantitative Tool For Analyzing Magmatic Volatiles

Hydrothermal experiments were conducted on high-silica (73-75 wt% SiO₂), fluid-saturated melts at 844-862°C and ca. 50 MPa using crushed glass of the Los Posos rhyolite.  Water and salts including NaCl, KCl, Ca(OH)₂, and CaHPO₄ and HCl were added proportionally to the experiments to restrict the variability of the aluminosity of the melt.  The Durango apatite, which contains 3.53 wt% F and 0.41% Cl, was added as “seeds”<5µm in diameter to stimulate apatite growth during the experiments.  Samples were loaded into gold capsules and run in cold-seal pressure vessels for durations of 286-1008 hours.  Temperature was cycled at ±20˚C to enhance apatite crystallization.  Electron microprobe analyses of run-product glasses and embedded apatite grains support calculation of a range of partition coefficients ( = wt% Cl in apatite/wt% Cl in melt) of 4.7 to 15.9.  The molar ratio of Cl in experimental apatites, or XCl, ranges from 0.19 to 0.56, while XF ranges from 0.08 to 0.63.  The computed values for XOH range from 0.24 to 0.38.  We find that normalizing XCl to XOH of apatites dramatically improves the precision when using apatite compositions to model Cl contents of melts. We compare our Los Posos rhyolite experiments with published data on 50 MPa rhyodacite experiments and find that Cl partitioning is significantly different in each system.Given the importance of chlorine in fluid equilibria, ore transport, and magma evolution, applications of apatite as a proxy for Cl contents in melts are unbounded.  It is found that in order to accurately use the volatile composition of natural and synthetic apatites to calculate the volatile composition of melts in felsic systems, several chemical factors, including wt% SiO₂ and the aluminosity/alkalinity of melts, should be incorporated as parameters to enhance relevant modeling.  This allows geochemists to place better constraints on processes associated with crystallizing Cl-bearing magmatic systems.

Sabrina Berger (UC Berkeley)

Mentor: Jana Grcevich

Evaluating GALFA-HI Survey Data For Potential Nearby Dwarf Galaxies

Many more satellite galaxies to the Milky Way have been predicted with cosmological simulations than have been observed. With data from the Galactic Arecibo L-Band Feed Array Survey Data Release Two, there is now an even broader sky to examine for interesting objects or possibly even undiscovered ultra-faint satellite dwarf galaxies. After careful analysis of many of the data cubes from this data release, over 60 interesting objects have been found so far. The relatively low range of radial velocity relative to the local standard of rest (V_LSR) of the survey mostly constrains the objects’ distances to within the vicinity of our galaxy.  The large range of velocity width, V_LSR and peak column densities found for the objects altered the goal of the project — since many of the interesting objects did not appear to be galaxy-like. Possible dwarf galaxies found within the data had large velocity widths (at least 20 km/s) and peak column densities of order of magnitude of at least 10¹⁹ (approximately that of the ultra-faint dwarf satellite, Leo T). However, there may be some association with galactic inflow and outflow and these clouds, but there is not strong evidence either way for this so far. An example of a compact cloud near an Algol-type eclipsing binary star system at a galactic longitude of 133.246 degrees and galactic latitude of -29.136 degrees was also found. Previous and future studies may also provide further evidence for a relationship between binary stars and compact clouds. With the coverage of the soon to be released optical survey, Panoramic Survey Telescope and Rapid Response System (Pan-STARRS), we may be able to examine some of these objects more thoroughly than ever before.

Candace Wygel (Skidmore College)

Mentor: Celine Martin

Lithium isotopic measurements in spodumene, lithiophilite, tourmaline, and beryl by LA-MC-ICP-MS

Lithium (Li) is an incompatible fluid-mobile element, which records metamorphic fluid-rock interactions in subduction zones. Most of the previous analyses performed on Li isotopes regarding subduction zone materials have been done on whole-rock. As metamorphic minerals tend to display chemical zoning, a new method coupling a laser-ablation microscope with a MC-ICP-MS was developed in Lamont Doherty Earth Observatory - Columbia University to measure Li isotopes in situ. It is called LA-MC-ICP-MS for Laser Ablation Multi-Collector Ion Coupled Plasma Mass Spectrometer. Measurements are conducted by LA-MC-ICP-MS in both low resolution (LR) and medium resolution (MR) to check if there is any interference on ⁶Li or ⁷Li. The δ⁷Li of a given sample is expected to be similar in both LR and MR, except in the case of interference. Lithium isotopes were measured in thirteen Li-bearing minerals from the mineral collection of the American Museum of Natural History. Samples include six tourmaline samples (two elbaites, one liddicoatite, and three schorls), one phosphate (lithiophilite), four pyroxenes (spodumene), one beryl, and one petalite. Lithium concentrations were measured by Laser Ablation Ion Coupled Plasma Mass Spectrometer (LA-ICP-MS). The amount of lithium ranges from about 40,000 ppm (lithiophilite) to about 500 ppm (schorl) in order to analyze a large range of Li concentrations in the minerals. The data obtained is compared to previously published data on Li isotopes in pyroxene, tourmaline, and phosphate (Martin, et al., 2015). Lithium isotopic analysis by LA-MC-ICP-MS gave significant differences in δ⁷Li values compared to previously published values, whether using the same analytical method or a different one. For example, the published δ⁷Li value of a lithiophilite sample is 8.9 ± 1.0 ‰ (Martin et al., 2015) whereas the measurements of the present study range from +17.0 ± 1.5 to +31.3 ± 1.5 ‰. Most of the various mineral species analyzed display a shift of about 7 ‰ between LR and MR, without any evidence of an interference peak. The amount of Li in the samples also did not have an effect on the consistency of the method. Therefore, it seems that LA-MC-ICP-MS is not a suitable method to measure Li isotopes in a large range of minerals (pyroxene, tourmaline, phosphate). Future analyses on Li isotopes should continue to be performed by SIMS.

John M. Christoph (College of William and Mary)

Mentor: Denton Ebel/Ellen Crapster-Pregont

Complementarity Relationships in the Trace Element Chemistry of CV type Meteorites 

The process of chondrule formation is a missing link in the present understanding of Solar System evolution from a chaotic protoplanetary nebula to stable planetary bodies.  Chondrites display a wide variety of components, including chondrules, calcium-aluminum-rich inclusions (CAIs), amoeboid olivine aggregates (AOAs), cryptocrystalline or glassy material, and fine-grained matrix.  Geochemical and petrological observations of chondrite components reveal that nebula and accretion processes in the early solar system were varied. Observations of the major element (Mg, Si, and Fe) bulk chemistry show complementary among components within a given chondrite. These major elements are enriched in individual chondrite components, but complement each other in such a way that bulk chondrite maintains its nearly solar composition. However, it is unclear how the less abundant refractory components, such as CAIs, contribute to maintaining complementarity.   Investigation into whether REE maintain complementarity yields insight into early solar system processes and formation history of chondritic components, including refractory inclusions where REE are particularly concentrated. Building upon previous work on the CO chondrites, this project examines the distribution of REE with the components of CV chondrites. CV chondrites contain comparatively large features but have similar overall abundances of each feature type to CO chondrites. The larger size of components also permits isolated analyses within the components and thus better determination of the REE contribution of exact mineral phases or component aspects. It is expected that the REE will complement each other among the chondritic components. These observed relationships are inconsistent with chondrule formation processes involving multiple reservoirs of source material, and require local formation and accretion of all measured components from the near-solar abundance protoplanetary nebula into the chondrites studied today.

Kate Storey-Fisher (Brown University)

Mentor: Ariyeh Maller

How To Define Dark Matter Halo Mass From Hydrodynamical Simulations

We present a comparison of common methods of defining dark matter halo mass. Using the cosmological hydrodynamical code RAMSES, we run a simulation containing dark matter and gas. We prevent star formation by introducing artificial pressure at a certain density to support the cold dense gas, eliminating the complications of feedback. This allows us to isolate the star-forming gas and analyze how this gas mass scales with halo mass. We use three common definitions of the halo mass, M₅₀₀, M₂₀₀, and M_vir, and we also compare using the total dark matter mass to including only bound dark matter particles. We find that M_vir produces halos in which the fraction of baryons to dark matter correlates most closely with the cosmic baryon fraction, Ω_b/Ω_dm, giving a mean error of ∼ 15%. We also determine that using the total dark matter mass produces halos with a baryon fraction significantly closer to Ω_b/Ω_dm, reducing the error by ∼50% compared with using only bound particles. However, none of the definitions consistently reproduce the cosmic baryon fraction in halos to better than ∼15%. This is important for populating halos in dark-matter only simulations with galaxies, in which Ω_b/Ω_dm is used to determine the gas and stellar mass in a halo. We conclude that a mass-dependent halo definition may better represent Ω_b/Ω_dm and give more accurate results in simulation analyses.